Source Preference in the Wild: How LLM Agents Favor Items by Source, and How to Reduce It
Factual evidence
What the source reports
Research across 12 agent models reveals persistent source bias in LLM search, favoring specific platforms over identical alternatives.
Inspect the evidence
- Inclusion basis
- Enterprise AI
- Publisher and source type
- arXiv cs.CL — Computation and Language · RESEARCH
- Published by source
- 5 October 2026
- Collected by OneBench
- 6 Oct 2026, 03:01 UK
Stored source excerpt
arXiv:2610.03195v1 Announce Type: new Abstract: As LLM agents decide on users' behalf which product to buy, which hotel to book, or which paper to cite,…
Short excerpt from the collected text, not the full source. Use the source link to read it in context.
The factual summary is a OneBench synthesis, not a quotation or independent verification. Collection time is not publication time. Open the source for its full context; related reporting can share the same underlying announcement.
OneBench interpretation
Institutional assessment
So what
Autonomous agents used in financial product recommendations or research retrieval risk introducing non-neutral bias, creating undisclosed suitability and conduct risks.
Do what
Review source-selection bias and retrieval weighting when evaluating consumer-facing or advisory agentic workflows with the model risk team.